Treffer: Data-driven Insights Machine Learning Approaches for Netflix Content Analysis and Visualization

Title:
Data-driven Insights Machine Learning Approaches for Netflix Content Analysis and Visualization
Contributors:
Hochschule Bonn-Rhein-Sieg University of Applied Science, 53754, Sankt Augustin, Germany., Sapienza University of Rome, Piazzale Aldo Moro, 5 – 00185, Italy., University of Florence, Italy., University of Naples Federico II = Università degli studi di Napoli Federico II (UNINA)
Source:
Journal of Engineering Research and Reports. 27(4):278-290
Publisher Information:
CCSD, 2025.
Publication Year:
2025
Original Identifier:
HAL: hal-05031761
Document Type:
Zeitschrift article<br />Journal articles
Language:
English
ISSN:
2582-2926
Relation:
info:eu-repo/semantics/altIdentifier/doi/10.9734/jerr/2025/v27i41471
DOI:
10.9734/jerr/2025/v27i41471
Accession Number:
edshal.hal.05031761v1
Database:
HAL

Weitere Informationen

This paper looks at Netflix's strategic application of machine learning and data analytics to improve user involvement, maximize content strategy, and keep its leadership in the cutthroat streaming market. This research reveals important patterns in Netflix's content library including the geographical distribution of content creation, content classification by rating, and changing watching habits over time by use of exploratory data analysis (EDA) and sophisticated visualization tools like Python and Tableau. According to the study, Netflix's content and user data is mostly produced by the United States (36.6%) and India (24.1%), followed by other nations including Japan, France, and Canada albeit in lesser but noteworthy proportions. Moreover, a high inclination for adult audience material is clear: 43.0% of TV series rated "TV-MA" and 33.7% of movies categorized under the same grade. Using clustering and regression among other machine learning methods, content success is predicted and audience preferences are analyzed, therefore illuminating the impact of particular genres and directors on audience trends. Content additions show a spike in output between 2014 and 2020, with the United States keeping leadership as nations like South Korea and India become more well-known via a time-series study. Correct data integrity guarantees by data preprocessing—including null value analysis—allows correct insights. With genres like "Stand-Up Comedy" and "Dramas, International Movies" rising as top categories, the report also emphasizes Netflix's dependence on prominent filmmakers and genre-specific content initiatives. This work shows how data-driven decision-making impacts Netflix's content acquisition and recommendation system by combining visualizing with machine learning. Future studies should investigate geographical variances, sentiment analysis, and predictive modeling to better grasp audience involvement techniques and streaming industry dynamics.